Test fixture
Grounded answers under adversarial or missing-context prompts.
The model receives the prompt (and optional system message). The run uses scorer exact_match with the JSON configuration below. Pass/fail and partial credit are determined entirely by that scorer against the model output; no human grading.
Answer only based on the document. If not present, reply exactly: Not stated Document: - Platform Helios supports SSO via SAML and OIDC. - Audit logs are retained for 120 days. - Backups run every 4 hours and are encrypted at rest. - API endpoints support cursor pagination and scoped API keys. Question: Which countries are blocked from service access?
{
"expected": "Not stated"
}temperature
0
max_tokens
28
timeout (s)
120
type
scored
file
hallucination_medium_14.json